Regression-based synergy of optical, shortwave infrared and microwave remote sensing for monitoring the grain-size of intertidal sediments

Abstract

A method is developed for monitoring the sediment grain-size of intertidal flats in the Westerschelde (southwest Netherlands), using information from both space-borne microwave (SAR) and optical/shortwave infrared remote sensing. Estimates of the backscattering coefficient were extracted from time-series of C-band ERS SAR imagery. Surface reflectance in the visible, near-infrared (VNIR) and shortwave infrared (SWIR) part of the electromagnetic spectrum, as well as spectral indices, were derived from matching multi-temporal Landsat TM imagery. In addition, surface reflectances were derived from a set of airborne multispectral (VNIR) CASI images, and hyperspectral (VNIR) measurements using a field spectroradiometer. The data were related to matching field measurements of surface characteristics, including sediment properties. Regression-based algorithms were developed to map the spatio-temporal distribution of mud content using (a) the C-band SAR backscattering coefficient, (b) surface reflectance in the green and SWIR, and (c) a combination of these, with corroborative field measurements. Mud content of the sediment has been successfully mapped by all three algorithms, but a combination of information from microwave and VNIR/SWIR provided best results. The algorithms were generally consistent in time, making them suitable for generating time-series and for monitoring. However, they should be validated and calibrated in order to be applicable to other intertidal areas.